LianShuaiLong

how can i load the trained-model

2

hello, i finetune the pre-trained inception_v4 model with 3 classes,by model=make_model('inception_v4',3,pretrained=True) , and saved the trained-model as 'model.pth' by torch.save(model.state_dict(),'model.pth') but when i restored the model by model = make_model('inception_v4',3,pretrained=False)...

cannot import name 'dataclass_transform' from 'typing_extensions'

1

follow the (inference pipeline notebook](https://github.com/borisdayma/dalle-mini/blob/main/tools/inference/inference_pipeline.ipynb) i setup dependencies and libraries: `git clone https://github.com/huggingface/transformers.git` `cd transformers` `python setup.py build ` `python setup.py install` `git clone https://github.com/borisdayma/dalle-mini.git ` `cd dalle-mini` `python setup.py...

hi, when i run "python main.py --exploiter" ,it seems that the process will run forever? how can i stop the running process correctly and get the best result currently?

hi,first of all, i am insterested in your work.However, error: sqlite3.InterfaceError: Error binding parameter 3 - probably unsupported type. occurs when i try to run ' python main.py --exploiter' in...

how can i retrian the efficientnet model with tensorflow framework? Can you give me an example?

请问一下，将模型转成trt的过程中，如何支持任意大小的输入？

Doubts about KL loss

6

Here are your codes when calculate KL loss: ``` log_qy = F.log_softmax(logits, dim = -1) log_uniform = torch.log(torch.tensor([1. / num_tokens], device = device)) kl_div = F.kl_div(log_uniform, log_qy, None, None, 'batchmean',...

thanks for your brillant implement of efficientnet transfer learning! i have some trouble in follow your example: when i run the example with gpu ,it cannot start trainning,can you help...

[Bug]: preprocess is too slow and always cause sd_webui error

1

### Is there an existing issue for this? - [X] I have searched the existing issues and checked the recent builds/commits of both this extension and the webui ### What...